AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Dynamic MDETR: A Dynamic Multimodal Transformer Decoder for Visual Grounding

Fengyuan Shi, Ruopeng Gao, Weilin Huang, Limin Wang

TPAMI 2023 · Vision-language and multimodal › Multimodal LLMs · approach: Training-free token reduction for MLLMs · 74 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
decouple grounding into encoder plus text guided decoder (all points) (ablation)TransVG · RefCOCOg val-utop-1 accuracy: 67.66 -> 69.12Improved
add 2D adaptive sampling (36 points) to the text guided decoder (ablation)Dynamic MDETR · RefCOCOg val-uaccuracy and GFLOPS: 69.12 -> 69.20; 4.10 -> 2.23 GFLOPSImproved
add text guided decoding on top of 2D adaptive sampling (ablation)Dynamic MDETR · RefCOCOg val-utop-1 accuracy: 67.89 -> 69.20Improved
use learnable initial sampling instead of non-learnable grid or uniform (ablation)Dynamic MDETR · RefCOCOg val-utop-1 accuracy: 67.77 -> 69.20Improved
use dynamic 2D adaptive sampling instead of static uniform sampling (ablation)Dynamic MDETR · RefCOCOg val-utop-1 accuracy: 66.81 -> 69.20Improved
add language guidance to the sampling query (ablation)Dynamic MDETR · RefCOCOg val-utop-1 accuracy: 68.32 -> 69.20Improved
progressively reduce sampled points (64-32-16) instead of constant 36 (ablation)Dynamic MDETR · RefCOCOg val-utop-1 accuracy: 69.20 -> 69.34Improved
use only 4 sampled points (1% of points) (ablation)Dynamic MDETR · RefCOCOg val-utop-1 accuracy: 0.22% dropNo clear effect
use text guided decoding instead of average pooling of sampled features (ablation)Dynamic MDETR · RefCOCOg val-utop-1 accuracy: 67.89 -> 69.20Improved
use text guided decoding instead of self attention decoding (ablation)Dynamic MDETR · RefCOCOg val-utop-1 accuracy: 68.30 -> 69.20Improved
use language queries instead of MDETR object queries (ablation)Dynamic MDETR · RefCOCOg val-utop-1 accuracy: 66.81 -> 69.20Improved
use encoder-decoder with 2D sampling instead of encoder-only 1D attention-based samplingTransVG · RefCOCOg val-utop-1 accuracy: 67.67 -> 69.20Improved

Later papers that compared with this method

2 later accepted papers report Dynamic-MDETR as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy